Expression and Prognostic Significance of Metastasis-Associated Protein 1 in Gastrointestinal Cancer.

Background Metastasis-associated protein 1 (MTA1) has been considered as a transcriptional regulator, which is significantly related to the prognosis in various types of tumors. However, whether MTA1 is a potential prognostic index of gastrointestinal cancer (GIC) remains controversial. The current meta-analysis was performed to evaluate the role of MTA1 expression in the prediction of the clinicopathological features and survival in GIC cases. And the results of gastric cancer were verified by immunohistochemistry (IHC). Methods Eligible studies assessing the relationship between MTA1 and GIC by IHC were searched in the PubMed, Cochrane, Ovid, Web of Science and CNKI databases by various search strategies. The STATA 16.0 software was applied to gather data and to analyze the potential relationship between MTA1 and GIC. The expression level of MTA1 was examined in 80 GC samples by IHC assay. SPSS 20.0 was applied for statistical analysis, and the survival curves were calculated by the Kaplan-Meier method. The data of 95% CI was displayed as "[a-b]". Results According to the meta-analysis, the expression level of MTA1 was tightly associated with the tumor size (OR=1.82 [1.16-2.84], P=0.009), tumor tissue differentiation (OR=1.71 [1.24-2.37], P=0.001), depth of invasion (OR=3.12 [2.55-3.83], P<0.001), lymphatic metastasis (OR=2.99 [2.02-4.43], P<0.001), distant metastasis (OR=4.66 [1.13-19.24], P=0.034), TNM stage (OR=4.28 [2.76-6.63], P<0.001). In addition, MTA1 played the negative effects in 1- (RR=2.48 [1.45-4.25], P=0.001), 3- (RR=1.66 [1.30-2.11], P<0.001) and 5-year (RR=1.73 [1.37-2.20], P<0.001). Study in subgroup, grouped by language and tumor type, we reached similar conclusions. Further validation by IHC yielded similar conclusions. Tumor size (P=0.008), lymph node metastasis (P=0.007) and distant metastasis (P=0.023) significantly accompanied with higher expression of MAT1 in GC cases. Besides, the expression level of MTA1 was statistically significantly correlated with OS in GC cases (HR=2.061 [1.066-3.986], P=0.032), which suggested that MTA1 might be an independent prognostic marker for GC. Finally, we verified the correlation between the expression level of MTA1 and prognosis of GC in 80 GC samples. Conclusions MTA1 is tightly associated with metastasis-related factors and may constitute a promising prognostic factor of GIC.